ABC analysis: directing attention across an uneven catalogue

Choose the ranking criterion consciously

Annual consumption value is the conventional basis, but it is not always the right one. Ranking by gross margin directs effort at profitability rather than turnover; ranking by unit value highlights where capital is tied up; ranking by pick frequency addresses handling effort. Different questions call for different rankings, and applying a classification built for one purpose to a decision it was not designed for is where most misuse originates.

One dimension is not enough

A high-value item with steady demand and a high-value item that sells sporadically need different treatment despite sharing a class. Overlaying a behavioural dimension — predictability, criticality to production, or shelf life — produces a grid rather than a list, and each cell of that grid maps to a coherent policy: tight triggers and frequent review here, generous buffers and infrequent attention there, order-on-demand somewhere else.

Translate class into concrete rules

A classification that does not change behaviour is decoration. It should determine review frequency, service target, replenishment model, counting cadence, and whether a human ever looks at the item. The most valuable class typically justifies individual attention and tighter buffers; the tail justifies simple rules, longer order cycles and periodic culling. Making these consequences explicit is what converts the analysis into an operating policy.

Reclassify on a cycle and watch the movers

Items migrate between classes as they launch, mature and decline, and a classification left static will apply yesterday's importance to today's catalogue. Beyond the periodic refresh, the interesting output is the movement list: items that have fallen sharply are candidates for buffer reduction or retirement, and items climbing quickly may be running on parameters set when they were marginal.

Frequently asked questions

Where should the class boundaries be drawn?

Where the curve actually bends for your catalogue, not at conventional cut-offs borrowed from a textbook. The purpose is to separate groups that will be managed differently, so the boundary is a management decision informed by the distribution rather than a statistical constant.

Should low-value items simply be eliminated?

Not automatically. Tail items can be essential spares, complete a range a customer buys as a whole, or serve a strategic account. The classification identifies where to ask the question; the answer requires commercial context the ranking does not contain.

How does the classification affect stock counting?

It sets the cadence. Items whose value or movement makes an error expensive are counted more often, while the tail is counted rarely, which spreads the effort across the year and finds discrepancies where they matter rather than treating every line as equally worth checking.